博碩士論文 etd-0707107-004933 詳細資訊


[回到前頁查詢結果 | 重新搜尋]

姓名 林宏陽(Hung-yang Lin) 電子郵件信箱 E-mail 資料不公開
畢業系所 資訊管理學系研究所(Information Management)
畢業學位 碩士(Master) 畢業時期 95學年第2學期
論文名稱(中) 運用機率架構於網格環境之動態資源規劃
論文名稱(英) A Probability-based Framework for Dynamic Resource Scheduling in Grid Environment
檔案
  • etd-0707107-004933.pdf
  • 本電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
    請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
    論文使用權限

    電子論文:校內一年後公開,校外永不公開

    論文語文/頁數 英文/60
    統計 本論文已被瀏覽 5349 次,被下載 7 次
    摘要(中) 近年來由於網格運算的盛行,有越來越多的研究針對如何分配分散的網格資源給工作流程中的每個任務提出不同的解決方法,而過去大部分的研究主要是著重在如何降低完成一個工作流程所花費的時間,而且他們大多是將工作流程中的每個任務在不同的資源執行所需的時間當成是固定的常數來處理,因此在本篇論文,我們提出一個以機率為架構的網格資源分配模型,將不同的任務在不同的資源執行所需要花費的時間當成是一個常態分佈,因此每個任務在相對應的資源所要花費的執行時間變成了一個變數而不是常數,因此將一些不確定性納入考量,我們的目標是希望能夠動態的分配資源給工作流程中的任務,進而使整個工作流程能在使用者所期望的時間內完成的機率最大化,也就是找出一個能符合使用者所希望完成的時間內完成工作流程的資源分配,我們提出了三個演算法來動態的處理網格資源分配,包括integer linear programming, the max-max heuristic and the min-max heuristics,並也提出兩個將執行時間當成常數的方法做比較,為了提高實驗的可靠性,我們採用真實的一個工作流程應用來進行實驗,並以模擬的方式進行各種不同資源的環境,最後我們得到的結果顯示在大部份的形況下,the min-max heuristics 表現的比其他方法好。
    摘要(英) Recent enthusiasm in grid computing has resulted in a tremendous amount of research in resource scheduling techniques for tasks in a workflow. Most of the work on resource scheduling is aimed at minimizing the total response time for the entire workflow and treats the estimated response time of a task running on a local resource as a constant. However in a dynamic environment such grid computing, the behavior of resources simply cannot be ensured. In this thesis, thus, we propose a probabilistic framework for resource scheduling in a grid environment that views the task response time as a probability distribution to take into consideration the uncertain factors. The goal is to dynamically assign resources to tasks so as to maximize the probability of completing the entire workflow within a desired total response time. We propose three algorithms for the dynamic resource scheduling in grid environment, namely the integer programming, the max-max heuristic and the min-max heuristic. Experimental results using synthetic data derived from a real protein annotation workflow application demonstrate that the proposed probability-based scheduling strategies have similar performance in an environment with homogeneous resources and perform better in an environment with heterogeneous resources, when compared with the existing methods that consider the response time as a constant. Of the two proposed heuristics, min-max generally yields better performance.
    關鍵字(中)
  • 機率
  • 網格
  • 工作流程
  • 資源規劃
  • 關鍵字(英)
  • Probability
  • Grid
  • Workflow
  • Resource Allocation
  • 論文目次 CHAPTER 1 - Introduction 1
    CHAPTER 2 - Literature Review 4
    2.1 Clustering computing to grid environment 4
    2.2 General grid environment 5
    2.2.1 Static resource scheduling 6
    2.2.2 Dynamic resource scheduling 9
    2.3 Comparison 10
    CHAPTER 3 - The System Model 14
    3.1 System architecture 14
    3.2 Autonomy of grid resources 16
    CHAPTER 4 - Dynamic Scheduling Algorithms 18
    4.1 Method to specify individual threshold 18
    4.2 Scheduling algorithm 20
    4.2.2 Integer linear programming (ILP) scheduler 21
    4.2.2 Max-max and min-max schedulers 22
    CHAPTER 5 - Performance Evaluation 27
    5.1 Simulating Environment Settings 28
    5.2 Experimental Results 30
    5.2.1. Success rates in the homogeneous environment 31
    5.2.2 Success rates in the heterogeneous environment 32
    5.2.3. Scheduling Strategies Comparison 34
    5.2.3.1 min-min v.s. max-min 34
    5.2.3.2 min-max v.s max-max 36
    5.2.3.3 min-max v.s. min-min 44
    CHAPTER 6 - Conclusions 48
    References 49
    參考文獻 A. O’Brien, S. Newhouse and J. Darlington. (2004.). Mapping of scientific workflow within the e-protein project to distributed resources. UK e-Science all Hands Meeting, Nottingham, UK,.
    Afzal, A., Darlington, J., & McGough, A. S. (2006). Stochastic workflow scheduling with QoS guarantees in grid computing environments. Fifth International Conference on Grid and Cooperative Computing (GCC'06), Changsha, China. , 0 185-194. from http://doi.ieeecomputersociety.org/10.1109/GCC.2006.89
    Blythe, J., Jain, S., Deelman, E., Gil, Y., Vahi, K., Mandal, A., et al. (2005). Task scheduling strategies for workflow-based applications in grids. , 2 759-767 Vol. 2.
    Braun, T. D., Siegel, H. J., Beck, N., Bölöni, L. L., Maheswaran, M., Reuther, A. I., et al. (2001). A comparison of eleven static heuristics for mapping a class of independent tasks onto heterogeneous distributed computing systems. Journal of Parallel and Distributed Computing, 61(6), 810-837.
    Cao, J., Jarvis, S., Saini, S., & Nudd, G. (2003). GridFlow: Workflow management for grid computing. Cluster Computing and the Grid, 2003.Proceedings.CCGrid 2003.3rd IEEE/ACM International Symposium on, , 198-205.
    Daniel Paranhos da Silva, Cirne, W., & Francisco Vilar Brasileiro. (2003). Trading cycles for information: Using replication to schedule bag-of-tasks applications on computational grids. Euro-Par 2003 Parallel Processing, Klagenfurt, Austria. 169-180.
    Duan, R., Prodan, R., & Fahringer, T. (2006). Run-time optimisation of grid workflow applications. Grid Computing, 7th IEEE/ACM International Conference on, Sci., Innsbruck Univ. 33-40.
    Foster, I., & Kesselman, C. (1998). The grid: Blueprint for a new computing infrastructure.Morgan Kaufmann Publishers Inc. San Francisco, CA, USA.
    Foster, I., Kesselman, C., & Tuecke, S. (2001). The anatomy of the grid: Enabling scalable virtual organizations. International Journal of High Performance Computing Applications, 15(3), 200-222.
    Galstyan, A., Czajkowski, K., & Lerman, K. (2004). Resource allocation in the grid using reinforcement learning. 1314-1315.
    Garey, M. R., & Johnson, D. S. (1979). Computers and intractability: A guide to the theory of NP-completeness.WH Freeman & Co. New York, NY, USA.
    Mandal, A., Kennedy, K., Koelbel, C., Marin, G., Mellor-Crummey, J., Liu, B., et al. (2005). Scheduling strategies for mapping application workflows onto the grid. 14-thIEEE Symposium on High Performance Distributed Computing (HPDC14), Research Triangle Park, NC,U.S.A. 125–134.
    Nudd, G., Kerbyson, D., Papaefstathiou, E., Perry, S., Harper, J., & Wilcox, D. (2000). Pace—A toolset for the performance prediction of parallel and distributed systems. International Journal of High Performance Computing Applications, 14(3), 228-251.
    Patel, Y., Mcgough, A. S., & Darlington, J. (2006). QoS support for workflows in A volatile grid. Grid Computing, 7th IEEE/ACM International Conference on, Imperial Coll., London. 64-71.
    Spooner, D. P., Cao, J., Jarvis, S. A., He, L., & Nudd, G. R. (2005). Performance-aware workflow management for grid computing. The Computer Journal, 48(3), 347-357.
    Sutton, R., & Barto, A. (1998). Reinforcement learning: An introduction
    Vanderster, D. C., Dimopoulos, N. J., & Sobie, R. J. (2006). Metascheduling multiple resource types using the MMKP. Grid Computing, 7th IEEE/ACM International Conference on, Victoria Univ., BC. 231-237.
    Yu, J., Buyya, R., & Chen Khong Tham. (2005). Cost-based scheduling of scientific workflow applications on utility grids. E-Science and Grid Computing, 2005.First International Conference on, Melbourne Univ., Vic., Australia. 8-16.
    Yu, J., & Buyya, R. (2005). A taxonomy of workflow management systems for grid computing. Journal of Grid Computing, 3(3), 171-200.
    Zeng, L., Benatallah, B., Dumas, M., Kalagnanam, J., & Sheng, Q. Z. (2003). Quality driven web services composition. Proceedings of the 12th International Conference on World Wide Web, 411-421.
    口試委員
  • 鄭炳強 - 召集委員
  • 李偉柏 - 委員
  • 陳嘉玫 - 委員
  • 黃三益 - 指導教授
  • 口試日期 2007-06-22 繳交日期 2007-07-07

    [回到前頁查詢結果 | 重新搜尋]


    如有任何問題請與論文審查小組聯繫